Integrate DALL-E 3 for Contextual Image Generation

[SverreNystad]

We propose to allow users to upload an image and provide a text prompt, enabling the AI to generate additional imagery that complements or extends the original picture contextually. The primary goal is to expand the creative possibilities for users, allowing for more personalized and contextually rich image generation. This could significantly enhance marketing materials, creative projects, and other visual content generated through MarketingAI. This feature has been requested by several users who have used MarketingAI and is a valuable addition.

Use Cases:

• Marketing Campaigns: Creating extended visuals for advertising based on a brand's existing assets.
  • For Cogito we can add our logo and generate imagery around it.
• Creative Projects: Artists and creators can build upon initial ideas or sketches to create more elaborate scenes.

Acceptance Criteria:

• The system must allow users to upload an image file (JPEG, PNG) up to a specified size (e.g., 10 MB).
• Users should be able to enter a text prompt alongside the image upload.

[SverreNystad]

This already exists for DALL-E 2 and would be quite easy to implement, but not jet for DALL-E 3. We should wait for this functionality to be released and then implement it.